Abstract
The molecular weight distribution of epoxide resins made from bisphenol A and epichlorohydrin at high ratios of epichlorohydrin to bisphenol A are compared with the theoretically predicted distributions for two theoretical models: the “taffy” process A, the direct reaction of epichlorohydrin with bisphenol A; and the “taffy” process B, the self‐polymerization of a monoglycidyl ether of bisphenol A followed by postglycidylation. At high ratios of epichlorohydrin to bisphenol A, process B is shown to give more low molecular weight products than process A. Deviations of the experimentally measured distributions from the theoretically predicted distributions for high epichlorohydrin/bisphenol ratios are attributed to the higher reactivity of epichlorohydrin to the phenolic compared with the aryl glycidyl ether functional group. Preliminary kinetic data are presented using a modified gel chromatographic method which enables the separation of most of the intermediates formed in this reaction.